You can test the impact of assets on your Performance Max campaigns. The primary goal of asset testing is to find out how adding or modifying assets impacts your Performance Max campaign's performance. This helps you answer key questions like:
- What's the value of adding assets to my product-feed only Performance Max campaigns?
- What's the incremental conversion lift from adding videos to my Performance Max campaign?
With asset testing, you can run experiments to measure the performance uplift from:
- Adding text, image, and video assets to a product feed-only campaign.
- Adding videos to a campaign that already has other assets.
On this page
How it works
Unlike other experiment types that test 2 separate campaigns, asset testing for Performance Max campaigns uses a single campaign. Traffic is split within that campaign, which can reduce the learning period and deliver faster results.
Test adding assets to a feed-only campaign
- Control group: Your existing Performance Max campaign with a Merchant Center product feed and no additional assets.
- Treatment group: The same Performance Max campaign, but with the text, image, and video assets you've added for the test.
You can test the performance of 2 different sets of assets in a Performance Max campaign. Learn more About Performance Max optimization experiments: A/B testing assets.
Test the impact of adding videos
- Control group: Your Performance Max campaign serves to a portion of traffic without any video assets. Both user-uploaded and auto-generated videos are suppressed for this group.
- Treatment group: The same Performance Max campaign serves to another portion of traffic with your video assets included.
- Note: All other assets (like text and images) will continue to serve to both the control and treatment groups. This isolates the impact of the videos to measure their incremental value.
Instructions
Recommendations for optimal setup
Note: Text customization was formerly known as automatically created assets.
- Make sure you don’t change the settings of text customization, Final URL expansion, or video enhancement feature after you create the experiment.
- Don’t change any assets in the campaign after the experiment has started.
- Ensure that the assets you upload adhere to Google Ads policies.
Set up an experiment to test adding assets to a feed-only campaign
- Go to Experiments within the Campaigns menu
.
- To create a new experiment, select the plus button
from the “All experiments” tab.
- Under “What do you want to test?”, select Assets.
- Under “Choose a variable to test”, select Assets provided by you. Then, select “Performance Max” as your campaign type.
- Select “Assets for retail campaigns” as the experiment subtype.
- Click Select campaign to choose an existing feed-only Performance Max campaign for the experiment, then select the Asset Group you want to test.
- Define the traffic split percentage for "Control" and "Treatment" arms.
- Under the “Treatment arm” card, select Add additional assets. Only supported assets are expandable and editable.
- Select the appropriate asset type to add the assets. Selected assets are summarized in the treatment arm card.
- To rename your experiment, click on the “Experiment name” field.
- Select the “Start date” and “End date” drop-down to select the start date and end date accordingly. Currently you can only choose today as your “Start date”.
- Select Schedule.
Set up an experiment to test the impact of adding videos
- Go to Experiments within the Campaigns menu
- To create a new experiment, select the plus button
- Under “What do you want to test?”, select Assets.
- Under “Choose a variable to test”, select Assets provided by you. Then, select “Performance Max” as your campaign type.
- Select “Video” as the experiment subtype.
- Click Select campaign to choose an existing Performance Max campaign that contains video assets.
- Define the traffic split percentage for the "Control" and "Treatment" arms. The experiment will automatically suppress videos for the control arm and serve them for the treatment arm.
- To rename your experiment, click on the “Experiment name” field.
- Select the “Start date” and “End date” drop-down to select the start date and end date accordingly. Currently you can only choose today as your “Start date”.
- Select Schedule.
Reporting
You can find the performance of your control and treatment groups in your Experiment Report page. The results summary table will show if you have enough data to get conclusive results. Learn more about how to Monitor your experiments.
Take action
After the experiment is complete, you can apply the winning variation or end the experiment.
- Apply experiment: This action applies the changes from your treatment group to your base campaign.
- For an “Assets for retail campaigns” experiment, this will add the new assets to your campaign.
- For a “Video” experiment, this will confirm that your campaign continues to run with videos.
- End experiment: This action ends the experiment without applying any changes. Your campaign will revert to its original settings.
